A pilot survey of in-service home arsenic tracked in from chromated copper arsenate-treated decks.
According to the National Research Council Subcommittee on Arsenic in Drinking Water, a wide variety of adverse health effects, including skin, lung, and urinary bladder cancers and cardiovascular and neurological effects, have been attributed to chronic arsenic exposure, primarily from drinking water (National Research Council [NRC], 2001; World Health Organization, International Agency for Research On Cancer, 2004). The NRC estimated that a daily dose of 3[micro]g/L arsenic in drinking water would result in an excess mean risk of bladder and lung cancer of one case per 1,000 persons (NRC, 2001).
Chromated copper arsenate (CCA), which was introduced in the 1930s, has been used extensively as an insecticide and fungicide in lumber for outdoor use. Although the U.S. Environmental Protection Agency (U.S. EPA) and the U.S. wood-preservative industry agreed to withdraw CCA-treated lumber for use in most residential settings starting January 1, 2004 (U.S. EPA, 2003), CCA was used to treat 98% of lumber produced for residential uses as recently as 2001 and existing inventories of CCA-treated lumber continued to be available for consumer use after U.S. EPA's ban (Consumer Product Safety Commission [CPSC], 2003). Although some health effects have been found for chromium and copper, effects due to arsenic are thought to be much more severe (CPSC, 2003; Patch & Maas, 2006). Exposure to arsenic from CCA-treated wood is thought to occur in three different ways: source-to-hand-to-mouth ingestion, inhalation, and dermal absorption (Zartarian, Xue, & Dang, 2005). Children are considered to be at especially high risk of exposure to arsenic in CCA-treated lumber because of their frequent mouthing behavior and lower body mass (CPSC, 2003). Hemond and Solo-Gabriele (2004) estimated that three-year-old children would receive an average arsenic dose of 33 [micro]g/ day from contact with CCA-treated decks and play sets, mostly from contact with their hands and subsequent mouthing of the hands. Zartarian and co-authors (2005) estimated that children with decks have approximately twice the absorbed doses of arsenic than children without decks. Dang and co-authors (2003) estimated that the average excess lifetime risk of bladder and liver cancer for children who have contact with CCA-treated decks and play sets during the ages 2-6 was 4.2 x [10.sup.-5] for children in warm climates and 2.2 x [10.sup.-5] for children in cold climates.
The above estimates do not include exposure that might be tracked indoors from CCA-decks because little is known about the potential for arsenic to accumulate on indoor floors as a result of being transferred through track-in from CCA-treated lumber. Lawn-applied pesticides, however, were found to be transported and to accumulate indoors via track-in (Lewis, Formann, & Damann, 1994; Nishioka, Burkholder, Brinkman, & Gordon, 1996). Lead dust is also known to be transported via track-in on shoes (U.S. Department of Housing and Urban Development [HUD], 1995). Sterling and co-authors (1999) found that the lead concentrations using a wipe method had a strong correlation ([R.sup.2] = 25%) to the blood lead levels of children (aged six months to six years) living in the homes, finding that 68 mg lead/[ft.sup.2] corresponded to a blood lead level of 10 [micro]g/dl.
In this project, four houses with CCA-treated decks and one house without a CCA-treated deck were selected so that researchers were able to test real situations instead of attempting to simulate them. Results may be used to determine if larger-scale testing of indoor floors and studies linking existence of arsenic on indoor floors with risk to humans is warranted.
Test sites were chosen by obtaining volunteers who owned homes with a deck or a porch that had been built prior to 2003 with CCA-treated lumber and had not been treated in the past two years. Upon selection, homeowners were asked to fill out a questionnaire concerning frequency of use, cleaning practices, treatment history, and other relevant general habits of inhabitants and visitors. Characteristics of the sites are listed in Table 1. Each site was tested three times: April 4 or 12, May 7 or 8, and September 10 or 11. Sampling was not allowed to occur during a 48-hour period after any precipitation event. Two wipe sample sites and a hand sample were located near the entrance associated with the CCA-treated lumber. For sampling, a template was made by cutting out two 100 [cm.sup.2] squares and one 300 [cm.sup.2] (20 cm x 15 cm) rectangle from foam poster board. The templates were cleaned using 1% solution of trace-metal grade nitric acid, rinsed with water, and dried with Kim Wipes[TM]. The door template was then placed adjacent to the door or, if a rug was immediately adjacent to the door, adjacent to the side of the rug opposite the door. The placement of the template was measured from the walls and documented so that samples were taken from the same places for every seasonal sampling. Using clean 100 [cm.sup.2] square templates, another wipe sample was obtained somewhere in the center of the room and, for a control, a blank wipe sample was obtained in an area of untrodden floor in a corner of the room. The locations of these samples were also carefully documented in order to use these exact locations in all three sampling runs.
The wipe sampling was consistent with the HUD method (1995) for lead dust-wipe clearance testing using a 100 [cm.sup.2] template, Ghostwipe[TM], and 50 mL vial with four exceptions. First, the 100 [cm.sup.2] templates used were much smaller than the recommended sample area of 1 [ft.sup.2] (929 [cm.sup.2]); this made the three sample sites adjacent to the door more closely packed and thus more likely to receive similar amounts of foot traffic and more easily managed in a template (This does, however, comply with the American Society for Testing and Materials [ASTM] recommendations ). Second, because of the small area there was not enough space for the researcher to perform the wipe sampling using the suggested technique involving the palm. Instead, wipes were taken using the undersurface of the fingers excluding the thumb. Third, although sampling was done with the undersurface of the fingers rather than the finger tips, because the guidelines caution against using fingertips, a second side-to-side s-shaped pass was included in addition to the first side-to-side and the following top-to-bottom motion.
An extra fold in the wipe preceded this third pass. The last deviation was that arsenic-spiked wipe samples were not submitted blindly to the lab but were spiked in the laboratory. For the hand sample, the researcher applied a clean hand having a surface area of approximately 150 [cm.sup.2], and an approximate force of 2 kg, with routine hand movement on a 300 [cm.sup.2] area sliding from the middle to the right, then back to the left, then back to the center and then turning approximately 45 degrees to slide into the left corner and again for the right corner. The hand was cleaned in the same way as the templates, using dilute, trace metal grade acid, distilled water, and Kim Wipes[TM] with a blank sample being taken from the clean hand (using the same hand wiping method described below) prior to sampling to ensure that the hand was not a source of arsenic contamination. After the hand was applied to the floor surface, a Ghostwipe[TM] was laid across the hand, moistened with the dilute nitric acid solution, and a wiping motion was performed with the gloved hand using the same folding technique as was used on the floor samples. This wipe and its previous blank sample were each placed in new 50 mL vials.
After samples were obtained the wipes were put in 60 mL disposable digestion vials. Calibration standards of 5.0, 25.0, 50.0, 75.0, and 100.0 parts per billion (ppb) were made by adding appropriate volumes of a 1,000 parts per million (ppm) arsenic solution to clean ghost wipes in separate digestion vials. Quality control standards (60.0 ppb) were prepared the same way, using a second source of standard solution. Also, for each batch of samples, two batch matrix QCs were performed by digesting a wipe plus a soil standard reference material (SRM) with a known standard concentration of 105 [micro]g As/g (arsenic per gram). Blanks were prepared with a clean ghost wipe in a digestion vial. All standards and samples in a batch were prepared together using a hot acid digestion. To dissolve the wipes, 6.0 mL of concentrated nitric acid and 2.0 mL of 30% hydrogen peroxide were added to each vial. They were covered with disposable, ribbed watch glasses and digested following U.S. EPA SW-846 Method 3050B, by heating at 95[degrees]C on a SCP Science Digibloc 3000 hot block digester. When the volumes of the samples were reduced to approximately 5 mL, they were removed from the digester and allowed to cool. Another 3.0 mL of concentrated nitric acid and 1.0 mL of 30% hydrogen peroxide were added to each vial, and the samples were again heated until the volume was reduced to 5 mL. The samples and standards were removed from the heat, allowed to cool, then poured up to 50.0 mL volume with deionized water.
Arsenic concentrations in each sample were determined using a Thermo Elemental SOLAAR M6 Graphite Furnace Atomic Absorption Spectrometer (GFAA), following SW-846 Method 7060A. Each sample and standard was poured into a 2.0 mL vial and loaded into the autosampler. During the automated analysis, the autosampler obtained an aliquot of each standard and sample, as well as nickel nitrate matrix modifier, and deposited it into the furnace cuvette, where the solution was heated and the arsenic atomized to allow detection. The instrument was calibrated and analytical QC samples were performed following calibration, every 10 samples and at the end of the analysis. These consisted of a blank, 50.0 ppb continuing calibration verification (CCV) sample and two 60.0 ppb QCs with acceptable recovery for the CCVs and QCs being within 85% and 115%. Acceptable levels for the blanks were [+ or -] 2.5 ppb (half of the concentration of the lowest calibration standard of 5.0 ppb). The batch matrix QC had acceptable recoveries of 85%-115%. The method detection limit (MDL) was found by multiplying the standard deviation of seven wipe samples containing 5.0 ppb As by a Student t-value of 3.143 to give an MDL of 0.10 [micro]g As per 100 [cm.sup.2] (Protection of Environment, 2007).
Geometric means of the arsenic per 100 [cm.sup.2] values were found by taking the mean of the log of each (value + 0.1) over the three visits for each home and location, then reverse transforming by the Exp (mean logged value) - 0.1 (Table 2).
The Pearson correlation coefficient was calculated on the mean logged values between deck wipe samples and indoor samples (Table 3) at each location. All sample sites were used except the control, so four means were included in each correlation calculation. Deck arsenic wipes were highly significantly correlated with indoor arsenic concentration at most sample locations (Table 3).
To investigate the relationship between the hand samples and wipe samples, the mean of the geometric means of the two wipe sample locations for each home for every sampling date was taken and the ratio of this value to the mean of the geometric means of the hand sample locations for each home was calculated (Table 4).
Several interesting observations can be made from the geometric means in Table 2. First, the highest sampling hand blank measurement of 0.02[micro]g/100[cm.sup.2] was well under the MDL of 0.10[micro]g/100[cm.sup.2]. This observation indicated that the sampling hand washing procedure was effective. Arsenic was removed and measured on the hand sample wipe. The arsenic concentrations at control site D were also very low, with the highest at 0.03[micro]g/100[cm.sup.2] and with the mean at 0.01[micro]g/100[cm.sup.2], a tenth of the MDL. This indicates that the arsenic concentrations that were present in the other houses were a result of the associated CCA-treated decks. The deck wipes from sample site H were consistently the lowest of the four non-control sites. In addition to having the lowest dislodgeable arsenic measured on its deck, sample site H had the least deck-to-indoor activity. These factors combine to explain why the geometric mean indoor arsenic levels for sample site H were the lowest in every indoor location. Sample sites B and K were fairly similar at each location. Sample site O, which had a deck arsenic concentration of about half of sites B and K, had roughly half of the indoor arsenic concentrations associated with sites B and K across the board as well. Arsenic found on wipe samples from the middle of the room was less than half of that found on door wipes, indicating that tracked-in arsenic concentrations become greater the closer one gets to the door adjacent to the CCA-treated deck. The majority of arsenic concentrations from wipe samples taken from untrodden corners were below the MDL. Even though one would expect deck use patterns and frequency to produce variations in indoor arsenic concentrations, these data suggest that the primary factor is availability of dislodgeable arsenic on the outdoor deck. Another pattern that can be seen in the data from Table 2 is the apparent relationship between the amount of arsenic generally dislodged indoors to the amount of arsenic dislodged from the deck.
This connection is supported by the Pearson correlation coefficients from Table 3. The correlation coefficients are all close to one, and the fact that so many p-values are below or close to .05 with such a small number of homes is fairly strong evidence that levels inside are closely tied to available arsenic on the deck. The correlations in our study may overstate the strength of the relationship in general, however, because the deck in our study with the lowest dislodgeable arsenic concentrations also received the lowest amount of foot traffic going from the deck to the indoor floor surface.
The mean of the ratio of the indoor wipe samples to adjacent hand samples was 10.36, which was greater than the ratios that were found in another study for direct contact with CCA-treated lumber residual (Patch & Maas, 2006). There were not enough sample sites to infer about the efficiencies of each wipe method as it is associated with different surfaces. The fact that the ratios for the wood floor in sample site B and the carpet in sample site H are both lower than those for the linoleum in the rest of the sample sites, however, is interesting.
Accurately estimating the average dose a child would receive from contact with arsenic tracked indoors would involve factors such as time spent crawling or playing indoors, ability to transfer arsenic onto the hands from carpets and several other factors that were not studied here. Comparison with the average dose estimated by Hemond and Solo-Gabriele (HS) (2004) may give an approximate order of magnitude of exposure. In that analysis, they assumed that a three-year-old child usually has moist hands when contacting CCA structures and estimated that an average hand load of arsenic on a moist hand was 70 [micro]g As/(100 [cm.sup.2] hand surface) from six studies that measured the arsenic transferred to a moist hand from contact with 700 [cm.sup.2] or more of CCA-treated structure. Four of those studies also had estimates for the amount transferred to a dry hand, which was a mean of 14.1 times less than for the moist hand. The mean dry hand arsenic amount for the current study was 0.75 [micro]g in the entrance area. Additionally adjusting for the fact that the mean wipe arsenic for the entrances was 0.393 [micro]g compared to 0.138 [micro]g in the middle of the room, that the hand wipe arsenic reported in Table 1 was for 100 [cm.sup.2] compared to the smallest area reported in HS of 700 [cm.sup.2], and the fact that the hand size in this study was 150[cm.sup.2] compared to the hand area of 100[cm.sup.2] in HS, the daily dose corresponding to a similar amount of contact in the middle of the entrance room would be
0.75 [micro]g In H/70 [micro]g HS H x 0.138 [micro]g W/0.393 [micro]g W
x 700 [cm.sup.2] HS WA/100 [cm.sup.2] In WA x 14.1M/1D
x 100 [cm.sup.2] HS HA/150 [cm.sup.2] In HA x 33. 0[micro]g
= 6.13 [micro]g,
In = indoor,
H = hands,
W = wipe,
A = area,
M = moist, and
D = dry.
This may underestimate the daily dose received indoors relative to HS because children, especially toddlers, would be expected to have much more contact indoors than outdoors. This can also be compared with the expected dose from infants drinking the estimated mean amount of water of 0.4 L at the 10 [micro]g maximum contaminant level for arsenic in drinking water, which is 4 [micro]g/day (Hemond & Solo-Gabriele, 2004).
Dislodgeable arsenic from CCA-treated lumber is transferred (likely via track-in) to the indoor environment. The degree of contamination decreases as distance from the entrance associated with the CCA-treated deck increases. Also, the amount of dislodgeable arsenic present on CCA-treated decks may be a larger factor than the amount of foot traffic across it and into the home.
A crude estimate of the potential exposure to children from arsenic tracked in from CCA-treated decks suggests that it may be of the same order of magnitude of that from direct contact with CCA-treated structures and from drinking water. A larger sample size is needed, however, to provide a reliable estimate of the amount of arsenic typically found indoors and to fully examine the effects of various factors including the deck, traffic patterns, and indoor flooring on indoor arsenic concentrations.
Acknowledgements: The authors thank Anne-Marie Traylor and Diane Morgan.
American Society for Testing and Materials International. (2003). ASTM E1793-03, Standard specification for wipe sampling materials for lead in surface dust. In Annual Book of ASTM Standards Volume 4.12. West Conshohocken, PA: ASTM International.
Consumer Product Safety Commission. (2003). Petition to ban chromated copper arsenate (CCA)-treated wood in playground equipment (Petition HP 01-3). Retrieved April 14, 2009, from http://www. cpsc.gov/LIBRARY/FOIA/Foia01/petition/Arsenic.pdf
Dang, W., Chen, J., Mottl, N., Phillips, L., Wood, P., McCarthy, S., Lee, R., Helmke, M., Nelson, M., & Coon, K. (2003). A probabilistic risk assessment for children who contact CCA-treated play sets and decks. Washington, DC: U.S. Environmental Protection Agency, Office of Pesticide Programs.
Hemond, H.E., & Solo-Gabriele, H.M. (2004). Children's exposure to arsenic from CCA-treated wooden decks and playground structures. Risk Analysis, 24(1), 51-54.
Lewis, R.G., Formann, R.C., & Damann, D.E. (1994). Evaluation of methods for monitoring the potential exposure of small children to pesticides in the residential environment. Archives of Environmental Contamination and Toxicology, 26(1), 37-46.
National Research Council. (2001). Arsenic in Drinking Water: 2001 Update. Washington, DC: National Academy Press.
Nishioka, M.G., Burkholder, H.M., Brinkman, M.C., & Gordon, S.M. (1996). Measuring transport of lawn-applied herbicide acids from turf to home: Correlation of dislodgeable 2, 4-D residues with carpet dust and carpet surface residues. Environmental Science Technology, 30, 3313-3320.
Patch, S.C., & Maas, R.P. (2006). Arsenic release and exposure from CCA-treated lumber. In T. Townsend & H. Solo-Gabriele (Eds.), Environmental Impacts of Treated Wood (pp. 215-235). Coral Gables, FL: University of Miami Press.
Protection of Environment. 40 C.F.R. [section] 136 (2007). Retrieved April 13, 2009, from http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecf r&sid=0dc100fe83b33755c7b5b0e8016c1936&rgn=div9&view= text&node=40:184.108.40.206.220.127.116.11.2&idno=40
Sterling, D.S., Roegner, K.C., Lewis, R.D., Luke, D.A., Wilder, L.C., & Burchette, S.M. (1999). Evaluation of four sampling methods for determining exposure of children to lead-contaminated household dust. Environmental Research Section, A81, 130-141.
U.S. Department of Housing and Urban Development. (1995). Guidelines for the evaluation and control of lead-based paint hazards in housing. Washington, DC: Author.
U.S. Environmental Protection Agency. (2003). Response to requests to cancel certain chromated copper arsenate (CCA) wood preservative products and amendments to terminate certain uses of other CCA products. Federal Register 68(68), 17366-17372. Retrieved April 14, 2009, from http://www.epa.gov/EPA-PEST/2003/ April/Day-09/p8372.htm
World Health Organization, International Agency for Research on Cancer. (2004). Some drinking-water disinfectants and contaminants, including arsenic. IARC monographs on the evaluation of carcinogenic risks to humans, vol. 84. Lyon, France: Author.
Zartarian, V.G., Xue, J., & Dang, O.W. (2005). A probabilistic exposure assessment for children who contact CCA-treated play sets and decks: Using the stochastic human exposure and dose simulation model for the wood preservative exposure scenario. Retrieved April 14, 2009, from http://www.epa.gov/heasd/sheds/pdf/CCA_Final.pdf
Steve Patch, PHD
Corresponding Author: Steve Patch, Director of the Environmental Quality Institute, the University of North Carolina at Asheville, Environmental Quality Institute, One University Heights, Asheville, NC 28804. E-mail: email@example.com.
TABLE 1 Characteristics of Houses in Study House Deck Code Age (Yrs.) Area Entrance H >10 1583 No K 3.5 3308 Main O 10 2262 Back B >12.5 5780 Main D (1) 20 539 Main House No. Shoes Indoor Code Residents Surface H 1 No Linoleum & Wood K 2 No Linoleum & Carpet O 4 Yes Linoleum B 4 50% Wood D (1) 5 Yes Wood (1) Control site. Porch was made of concrete. TABLE 2 Geometric Means of Sample Sites (a) L Door R Door Home Deck Hand (b) Wipe Wipe B 19.66 0.10 0.51 0.38 H 3.60 0.03 0.17 0.11 K 21.62 0.11 0.57 0.75 0 11.18 0.06 0.27 0.38 D 0.01 0.00 0.03 0.00 Home Middle Corner Pre-Hand B 0.17 0.10 0.02 H 0.00 0.01 0.00 K 0.26 0.09 0.02 0 0.12 0.03 0.00 D 0.01 0.02 0.00 (a) In [micro]g arsenic [As] per 100 [cm.sup.2]. (b) Hand sample concentrations were transformed to [micro]g As per 100 [cm.sup.2] by dividing by three, giving an MDL of 0.03 [micro]g As per 100 [cm.sup.2]. The original MDL of 0.10 [micro]g As per 100 [cm.sup.2] applies to all other locations. TABLE 3 Pearson Correlation Coefficient on Mean Logged Values Between Deck Wipe and Each Inside Location Sample Location Correlation p-Value Hand 0.97 .027 L Door Wipe 0.96 .044 R Door Wipe 0.93 .069 Middle 0.99 .013 Corner 0.93 .075 TABLE 4 Ratios of Geometric Means of Door Wipe Samples to Geometric Means of Hand Samples Home Ratio B 8.84 H 9.27 K 12.12 O 11.21 Mean 10.36
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ADVANCEMENT OF THE SCIENCE|
|Author:||Sigmon, Cole; Patch, Steve|
|Publication:||Journal of Environmental Health|
|Date:||Jan 1, 2010|
|Previous Article:||Prevalence of community-associated Methicillin-resistant Staphylococcus Aureus in high school wrestling environments.|
|Next Article:||Exposure to video display terminals and risk of small-for-gestational-age birth.|